Hi,
As you probably know there is a significant amount of hive in the dhf (decentralized hive fund): hiveblocks.com/@steem.dao, 83 millions hive to be precise. At the moment these tokens cannot be used by the dhf because payouts are only done in HBD.
If you don't remember why those tokens are here, it's because it's mostly the steemit ninja mined stake, which was ment to be used for the development of hive, so during the hard fork 23, the fork that birthed hive, those tokens were moved from the steemit accounts to the closest thing to it's original purpose aka the hdf.
Since hf24 is nearing completion, and the next one is in quite a bit, it makes sense to start converting that money to HBD in order for it to be usable, the code has been done by @netuoso and I a while ago : https://gitlab.syncad.com/hive/hive/-/merge_requests/2 Initially we wanted to have it ready by hf23 so it would start converting straight away and have funds ready to fuel development now that steemit inc isn't there anymore. But we felt it would be a change that's a bit too big and didn't want to rush it.
The big question
How fast should we convert that money ? We shouldn't convert all of it at once because the dhf allows you to withdraw 1% of its funds per day, which means that assuming a bad actor managed to get a proposal to payout, he could steal almost 200k in one day, we are far from spending 200k per day in development costs so there's no point in taking the risk.
If you are afraid of the economical implication with the hbd debt ratio, we will ship a change at the same time to make the hbd in the hdf not count towards it until it's paid out (which makes sense when you think about it, that money is locked up, it's not "debt" until it's actually usable by someone)
The idea is to slowly convert it in a way so that it reinforces the natural inflation of the fund without making it grow too quickly.
Some numbers to help you calculate
Currently the fund receives more or less 1752 additional hbd per day from the inflation
83 370 205 hive needs to be converted which equals to roughly 20 258 000 USD
I will assume that HBD = 1$ because it's easier
We convert on a per day basis, so every day a percentage of that money will be converted to HBD and made available.
For instance, if we were to convert all of it over a period of 5 years it would add
20258000/(365*5) = 11 100 HBD per day to the dao, which means 111 more per day that are available.
The thing is that this stacks up quite quickly:
after 100 days over 17k hbd are available per day, a decent amount if you ask me.
you are all free to play with the simulation spreadsheet I made here : https://docs.google.com/spreadsheets/d/1QTkPDPsjS5cUSMycHE7hPxtkCXJg54mNDLAyFNrbR_M/edit?usp=sharing
Some ideas
So the obvious answer you might think of would be "okay let's increase the time then", here's the thing, 5 years is a lifetime in crypto, so imagine more.
So the solution might be to decrease the 1% of available funds per day, but then some of the current people being paid out now might lose their funding because there isn't enough money available. Perhaps we could have a percentage based on how much is currently in the hdf like 1% if less than 1m hbd, 0.5% if more, and decrease as the fund grows ?
Should we allow people to be paid out in hive ? (not a fan of this one tbh)
I am open to suggestions, hence this post. So what do you think ?
Ps: If you think "Aren't we getting along just fine with the current amounts ?" We are overspending already. About 1752 HBD comes to the dao every day via the inflation, but we are spending ~2616 HBD per day.
You can see clearly in @penguinpablo's post: https://peakd.com/@penguinpablo/hive-dao-stats-thursday-july-2-2020
Daily budget (1% of all HBD available in the DAO):
Point is, we don't want to be in a position where we can't fund hive hive development even though we can because it's locked.
One solution is to make the %/day a witness parameter with a ceiling. This would allow some flexibility for the future.
My vote is 5 years or more. The DHF is growing a lot already on its own. I would be agreeable to slightly longer or shorter but I wouldn’t be favorable to too much shorter.
Without Steemit we will have a lot more expenses as well. I also believe we as a community will be far more active in terms of building so that will likely require more as well.
I do believe the more important thing is making the daily limit a witness parameter with valid limits of 0.1-3% or something similar.
Sounds reasonable to me.
That's actually a great point, I like the idea of having a witness param, I feel like the rate is something we might want to change relatively often depending on current economics.
I think the initial 800,000 sold every month by Steemit Inc is a healthy approximate number going forward. So I would round it up to 12.5% yearly or 8 years.
I think there is an argument to make the accessible % a witness parameter (capped) as well. What if some day an urgent security audit is required and a down payment needs to be made. But I would only favor that if 17/20 are counted and lowest common denominator of those 17. If a potential can of worms is opened, it should be made as hard as possible to game.
This sounds like a bug. The 1%, or actually 1/24th since paid hourly, should not be exceeded if hard coded in the chain and be distributed among the top voted proposals. If one proposal earns only a share because of that and some drop out... code is law and 1% is codified.
Times like this could be another argument for the witness parameter if the current paid out proposals cover critical dev.
The return proposal should be the protection when we underspend only. Top votesd ranking at other times.
That's a good point !
it's not a bug, it's 1% of the total amount in the fund, not 1% of the inflation to the dao.
I think it's a good mechanism because the fund may fluctuate depending on external factors like donations / ninja mine conversion / inflation.
My bad re-bug. I deducted from that one quoted sentence about "overspending" without checking the actual current numbers. I should have better there and fully agree with the rest of your reply about that.
no problems :)
I vote for slowly! 5-10 years for sure. As others have noted, the implications of a rising Hive price on the DHF can be very favorable, and I actually look at that as a potential metric of how well the DHF funds are being spent. If DHF funds are used well, they should bring value to Hive and elevate price, which grows the DHF and creates a virtuous circle that feeds on itself. Flooding the DHF with these ancillary funds too quickly could greatly harm the efficiency of spending, and distort our perception of just how effectively we’re allocating funds.
Why not burn this ninja mining completely instead to keep going with this legacy problem. The fund is rising already naturally with the parameters set in place.
Read my last point, the fund is depleting.
Perhaps the answer to this is to manage the fund and funded proposals better. If the fund is still not adequate then look at the source of the funding again. Is it enough? Or is it just not actively managed?
Why not rid ourselves of this stake? Why not strive to be different?
Burning it would also burn 20 million dollars in market cap since this stake counts towards that I believe.
After the 8th that will go down to 2116 HBD a day unless we are going to pay 500 HBD a day for longer than 60 days to have more communication.
I think we can easily burn 50% of the Hive that was seized and at least all of it that was additionally seized from people supporting Steemit (that Hive was never promised for development). The rest can be spread out over 5 years.
The argument of do we need the development works both ways, but I'm sure you developers have a bias towards keeping it for paying developers since you are 10000X times more likely to get it than others.
That 500 per day is lovely isn’t it?
Yes but no, right now a very significant amount of the hive development is not being funded (and will be soon-ish) @blocktrades has 4 or 5 (can't remember) full time devs working on it, and I assume he'll make a proposal for it. Which will take a significant amount of money.
@justineh is still doing work and I would also assume she might do another proposal.
My point is that we'll probably sustain even more expenses.
I am obviously biased being funded myself, and I'm truly sad to not see more marketing/communication proposals. I review all the new proposals and vote on the ones I like, and I do think that technical proposals are easier to sell in terms of tangible results.
It seems like there is clearly more in store and going on behind the scenes. I guess we will have to see what kind of proposals come out before we can get an accurate picture of what kind of funding is required and the value of these proposals.
However if people feel entitled to proposal money it becomes a problem , as it should ideally be efficiently distributed or the blockchain is going to crash when the money runs dry.
Like the idea of a long term conversion!
Also we can put a burn proposal to control the DHF funds.
That's a good idea ! Probably not burn it, but have something like "return 4000 HBD per day to the hdf" higher up so that proposals get funded, but not too many funds are available at the same time would be ideal.
I think 5 years is a good time period. It's long enough to prevent it from being spent all at once, yet short enough to have a strong impact on progress. It's important to note that this doesn't mean the funds have to be spent in 5 years, it just means it will be possible to do so, if that's what stakeholders want to do.
I also agree that HBD haircut calculations shouldn't take into account HBD held by the hive.fund account, so we need a modification to those calculations as part of the release.
It's not valid for the different prices of HIVE!
Imagine HIVE at $1, and we already have seen an all-time high 1$ in the non-bull market.
If we keep the HIVE as it is for longer time, the chance of converting it to more HBD increases. (assuming price in the moon)
So I would say it's not bad to look for +10 years.
Well if we see a clear surge we could update the percentage, we can do that at every hard fork if we feel like it, or set it as a witness param.
I don't support the idea of putting something important in the witness params as it opens the window for a quick change. But I think we can change the percentage on the next hard forks.
What the hell is this supposed to be?
A bottomless pit for the devs to put continual downward pressure on the price?
Give the authors their 10% back, and make do with only 20m usd, eh?
How much more 'development' are we looking at?
Right now I can't use vessel because it isn't compatible with my version of linux.
I created an issue, and bupkiss.
I know that guy is busy, but it is what he is getting paid for.
I think folks are losing sight of the fact we are supposed to be building something valuable here, not just leaching off the suckers until the whole thing collapses as may have been the previous intentions of some.
When do we get something that makes the price go up?
I don't think the inflation to hdf, should be removed, we want it to be sustainable via the inflation, not fed via a finite pool that will run out at some point
I think it's understandable that @netuoso is not prioritizing an issue that affects a super small set of the users, (most people are on debian/ubuntu or windows/macOs) and instead working on new features.
I don't think that was ever the case, how do you make price go up if you have no marketing, no development, no api nodes or no exchanges relationships (I'm looking at the current proposals). For the longest time steem worked on people working for free/witness rewards and steemit inc who was selling ninja mine.
Which ment that most people working full time were at steemit inc, now that it's gone is no longer sustainable to just have people work for free / for witness rewards.
I think we are literally doing that, if you disagree I would love to hear your thoughts in a post on what you think we should be doing, I think we have enough devs but are lacking a proper marketing/onboarding effort. Someone proposed at some point to get a proposal to pay a communication agency to market hive, and provided this is done by someone I trust will do a good job, I'm all for that. He just never went through with it to the end.
I don't know enough to contribute much to help find a balanced solution to this rock and a hard-place situation... but my thought was if there was a way to layer the %. Like give the daily average but set a trigger in a condition to allocate more so there could be the flexibility needed to fund the projects as they come up. My second thought was that if 5 years is forever in Crypto... going long would incentivize sticking with it to the end of the process. Not that those working hard to build this over here need to be incentivized to make it to the end of the race... and it's not to the swift! I am with you about not paying out in Hive.
Thanks !
Thank you for all that you and the others do to make all this possible for us!
I think 5 years is a good time. As you already pointed out, 5 years in crypto is an eternity (Just think about it, Steem is not even 5 years old).
The funds don't have to be all spent at once and I think it's more risky to "underfund" the initial developement than to overfund it.
How easy or difficult would it be to change the conversion rate in the future? If it's easy, then maybe go with a long time (10 years or even more) with the thought that for the time being it's more than sufficient and if at any point in time it becomes insufficient, the conversion rate can be increased.
It's a one var change (so extremely easy), but it would require a hard fork unless it's a witness parameter, which means that theorically we could set it every 3-4 months (no that many hard forks per year)
Cool, so perhaps flexible enough even without a witness parameter. Although I wonder if setting a witness parameter will create more witness and community engagement with the DHF.
Delete the DHF
Very interesting analysis you do, I do not understand much of this, but if you would advise us to help solve this situation, it would be truly fantastic. Greetings
While Hive is young, it seems to make sense to spend as much as possible on development of projects that are deemed worthy. Technology tends to mature over time, so allowing as many young projects to receive funding is the model that incubator and Venture Capitalist's embrace.
Like businesses most tech projects will fail, but that is exactly why those who are successful at development and funding, know for success you support many projects and the strong survive and thrive.
I tend to see the Hive stakeholders as wanting some sort of guarantee on their investments and while I think it is unintentional, (meaning not evil or even greed driven) they have a strong bias toward a small group of people.
I'd be more excited about Hive in general if I saw more movement towards attracting and embracing more talent, more users and most importantly new developers. Otherwise it will continue to stagnant.
Interesting perspective. It seems to be a little stagnant indeed. There are some cool things that have happened but the newest ideas from new people, where are they? I think the hackathon was fun and there was some potential there but I haven’t seen much follow up to promote more projects people are working on.
Hopefully we can get past all this and continue to generate new ideas and grow the business here rather than just be another reddit.
I would argue for a longer time period than 5 years.
I’ll try to convey my idea from way back then when I was thinking about Steem inc’s failure.
I always thought: "Come on! We are in crypto. Why are you converting Steem monthly like fuckin retards? Why didn’t you sell more Steem when it was high in order for you not to be forced to sell Steem when it’s low. That’s how most of us (at least try to) operate in this environment."
So what if, just for a moment, we scratched all this "convert daily/weekly/monthly" attitude and act like traders which we all HAVE TO BE...and yes...even the Hive itself if it wants to succeed or gain/attract more value. We cannot get stuck in rigidly set parameters.
So what if we set some sort of a "limit convert". I know we are young and stuff but we ALREADY MISSED ONE OPPORUTINITY when Hive reached 1 dollar.
What if we set limit converts for example at 0,50; 0,80; 1,00 bucks. We could change (or EVEN VOTE!!!) on those limits as we would see fit...are we running out of DHF’s money (now seems to be the case)? Well then let’s try to convert at 0,25-0,3 cents some of our holdings. What if we suddenly stop paying some of those proposals for any reason (for example @blocktrades 1,000 bucks a day is running out) and we are good for a while? Well then we could change the ratio and wait for 0,5-1,0 depending on what we see fit and the DHF’s economical buffer we currently run on.
Those are my raw ideas several years old. I just came back from a forced crypto holiday due to the university. Maybe they are outdated I will have to check some numbers again:).
Burn at least 50% of STINC stake and 100% that was seized from other users.
That convert it to HBD in 5-10 years.
Current DAO is being like a Government contract, we are just throwing a lot of money to people (mostly "friends" of large stakeholders, while projects like dapplr are unable to get funded with 30 HBD/day), and some of them are not doing much work (at least there is no transparency). While STINC was doing the same, they were a private company. DAO is being paid by every single HIVE stakeholder.
E.g Vessel had last public commit on Github on May 7th and it's being funded with 150 HBD per day. Even for a senior dev, that's 2-3 hours of work per day. In 64 days, there was no public activity and almost 10k HBD paid.
Difference between government contracts and the hdf is that you don't have any direct power on it, all you can do is rage that nasa awarded another cost+ contract to boeing despite them constantly failing.
Here you can directly affect it by unvoting yourself, raising awareness or contact large stakeholders to get a proposal you feel shouldn't be paid out/isn't doing enough work outside of the pay threshold.